[0001] The present disclosure relates to circuits for switched-mode-power-supplies, and
in particular to circuits that have a controller, a current-limiter and a clamp-circuit.
[0002] According to a first aspect of the present disclosure there is provided a circuit
for a switched-mode-power-supply, wherein the switched-mode-power-supply is configured
to:
receive a current-control-signal; and provide an output-voltage based on the current-control-signal,
the circuit comprising:
a controller, configured to:
generate a control-voltage based on the difference between: (i) a sense-voltage, which
is representative of the output-voltage of the switched-mode-power-supply; and (ii)
a reference-voltage;
generate a target-current-control-signal based on the control-voltage, wherein the
target-current-control-signal is configured to adjust the current through the switched-mode-power-supply
in order to bring the sense-voltage closer to the reference-voltage;
a current-limiter configured to provide the current-control-signal as the target-current-control-signal
limited to a max-current-control-value; and
a clamp-circuit configured to set the control-voltage to a clamp-voltage-value when
the current-limiter provides the current-control-signal having the limited value of
the max-current-control-value.
[0003] The clamp-circuit can advantageously prevent or reduce drift of the control-voltage
when current limiting is used. In this way the control-voltage may not leave a normal
operating region, and normal operation can be resumed more quickly when a load subsequently
reduces to a lower level that does not require current limiting. This can beneficially
reduce the overshoot on the output-voltage to a level that would occur without current
limiting.
[0004] In one or more embodiments, the clamp-circuit comprises a voltage regulator. The
voltage regulator may be configured to regulate the control-voltage to the clamp-voltage-value
for all values of the sense-voltage and the reference-voltage that would cause the
target-current-control-signal to be set to a value that is equal to or greater than
the max-current-control-value.
[0005] In one or more embodiments, the controller comprises a transconductance-amplifier-control-block,
which may incorporate the functionality of the current-limiter and the clamp-circuit.
The transconductance-amplifier-control-block may be configured to: receive the control-voltage;
prevent the control-voltage from increasing when the current-control-signal has the
limited value of the max-current-control-value; and convert the control-voltage to
the current-control-signal.
[0006] In one or more embodiments, the transconductance-amplifier-control-block comprises
one or more of: a voltage amplifier, a bias-voltage-source configured to provide a
saturation-voltage; a clamp-transistor; and a first-mirror-second-transistor. The
voltage amplifier in combination with the clamp-transistor may be configured to regulate
the voltage drop across the first-mirror-second-transistor to the saturation-voltage
provided by the bias-voltage-source.
[0007] In one or more embodiments, the transconductance-amplifier-control-block comprises
a transistor, and optionally the transconductance-amplifier-control-block is configured
to prevent the control-voltage from increasing when the voltage drop across the transistor
exceeds a saturation-voltage of the transistor.
[0008] In one or more embodiments, the saturation-voltage is representative of the transistor
being in a saturated state of operation.
[0009] In one or more embodiments, the transconductance-amplifier-control-block is configured
prevent the control-voltage from increasing when a transistor in the transconductance-amplifier-control-block
is in saturation.
[0010] In one or more embodiments, the controller comprises: a transconductance-amplifier-control-block,
which may incorporate the functionality of the current-limiter and the clamp-circuit.
The transconductance-amplifier-control-block may be configured to: receive the control-voltage;
provide a non-zero over-current-signal when the current-control-signal has the limited
value of max-current-control-value, wherein the over-current-signal may be configured
to change the sense-voltage independently of the current through the switched-mode-power-supply;
and convert the control-voltage to the current-control-signal.
[0011] In one or more embodiments, the switched-mode-power-supply includes a target-output-voltage,
which may define a set-point for the switched-mode-power-supply, and the over-current-signal
is configured to change the target-output-voltage to a level such that the current
through the switched-mode-power-supply corresponds to the max-current-control-value.
[0012] In one or more embodiments, the clamp-voltage-value is a value of the control-voltage
before the current-limiter provides the current-control-signal with the limited value
of the max-current-control-value.
[0013] In one or more embodiments, the clamp-circuit is configured to clamp the control-voltage
to its current value when the current-limiter provides the current-control-signal
with the limited value of the max-current-control-value.
[0014] In one or more embodiments, the controller is configured to generate the control-voltage
by integrating the difference between: (i) the sense-voltage; and (ii) the reference-voltage.
[0015] In one or more embodiments, the target-current-control-signal is configured to adjust
the current through the switched-mode-power-supply by setting the peak current through
an inductor in the switched-mode-power-supply.
[0016] In one or more embodiments, the clamp-circuit is configured to provide a flag-signal
that is representative of whether or not the control-voltage is set to the clamp-voltage-value.
The circuit may comprise a controllable-block that is configured to be automatically
controlled based on the flag-signal.
[0017] In one or more embodiments, the switched-mode-power-supply comprises a buck converter,
a buck-boost converter or a flyback converter.
[0018] There may be provided an audio amplifier circuit comprising:
any circuit disclosed herein, wherein the clamp-circuit is configured to provide a
flag-signal that is representative of whether or not the control-voltage is set to
the clamp-voltage-value; and
a controllable-block that is configured to be automatically controlled based on the
flag-signal.
[0019] In one or more embodiments, the controllable-block comprises an amplifier. The amplifier
may be configured to automatically control an audio level of an output signal of the
audio amplifier circuit based on the flag-signal.
[0020] While the disclosure is amenable to various modifications and alternative forms,
specifics thereof have been shown by way of example in the drawings and will be described
in detail. It should be understood, however, that other embodiments, beyond the particular
embodiments described, are possible as well. All modifications, equivalents, and alternative
embodiments falling within the spirit and scope of the appended claims are covered
as well.
[0021] The above discussion is not intended to represent every example embodiment or every
implementation within the scope of the current or future Claim sets. The figures and
Detailed Description that follow also exemplify various example embodiments. Various
example embodiments may be more completely understood in consideration of the following
Detailed Description in connection with the accompanying Drawings.
[0022] One or more embodiments will now be described by way of example only with reference
to the accompanying drawings in which:
Figure 1 shows an example circuit diagram of a boost converter;
Figure 2 shows a boost converter with peak current mode control;
Figure 3 shows a plot of current versus time for the inductor current IL of the circuit shown in Figure 2;
Figure 4 shows the same waveforms as Figure 3, and also an inductor-current-with-error
IL,ERR signal;
Figure 5 shows the same current waveforms in peak current mode control as those of
Figure 4, but this time for error propagation with |S2|>|S1|;
Figure 6 shows a boost converter with peak current mode control and slope compensation;
Figure 7 shows the same current waveforms in peak current mode control as those of
Figures 4 and 5, and also (IPEAK - IS);
Figure 8 shows a boost converter with peak current mode control and parabolic slope
compensation;
Figure 9 shows various current waveforms for the circuit of Figure 8;
Figure 10 shows a boost converter with peak current mode control and parabolic slope
compensation, with a more detailed view of the switches and the current measurement;
Figure 11 shows a boost converter with peak current mode control, parabolic slope
compensation and a P-I controller for providing an output-voltage control loop;
Figure 12 shows a boost converter, which is similar to the boost converter of Figure
11, with peak current mode control and parabolic slope compensation, with further
details of an implementation of the P-I controller;
Figure 13 shows a boost converter, which is similar to the boost converter of Figure
12, with peak current mode control, parabolic slope compensation, an output voltage
control loop and also a current-limiter;
Figure 14 shows an example implementation of a second transconductance-amplifier-control-block,
which provides the functionality of both the second transconductance-amplifier-control-block
and the current-limiter of Figure 13;
Figure 15 shows a plot of inductor current IL versus time for a step-up and step-down in the load current, for the circuit of Figure
13;
Figure 16 shows another plot of inductor current IL versus time for a step-up and step-down in the load current;
Figure 17 shows two plots of output-voltage VBST versus time;
Figure 18 shows two plots of the control-voltage Vc versus time;
Figure 19 shows an example embodiment of a circuit that includes a switched-mode-power-supply;
Figure 20 shows an example embodiment of a transconductance-amplifier-control-block
that provides the functionality of the second sub-processing block, the current-limiter,
and the clamp-circuit of Figure 19;
Figure 21 shows three plots of output-voltage VBST versus time, including a plot for the circuit of Figure 19 / 20;
Figure 22 shows the inductor current IL during a load step-up and step-down, including a plot for the circuit of Figure 19
/ 20;
Figure 23 shows three plots of the control-voltage VC versus time, including a plot for the circuit of Figure 19 / 20;
Figure 24 shows a plot of the flag-signal of Figure 20.
Figure 25 shows an example embodiment of the transconductance-amplifier-control-block
of Figure 20, which provides the functionality of the second sub-processing block,
the current-limiter, and the clamp-circuit of Figure 19;
Figure 26 shows an example embodiment of a transconductance-amplifier-control-block,
which is similar to that of Figure 25;
Figure 27 shows another example embodiment of a circuit that includes a SMPS and a
controller; and
Figure 28 shows an example implementation of the second sub-processing block of Figure
27.
[0023] In modern battery operated portable electronic equipment, such as smart phones, high
efficiency boost converters are used for circuits that require a voltage above the
battery voltage. In a boost converter, a controller is used to regulate the boost
voltage to the required level. The amount of overshoot or undershoot on this boost
voltage after a load step is an important performance indicator for the speed and
quality of the controller. If the load or the converter itself has a maximum voltage,
any overshoot can decrease the margin between maximum and nominal boost voltage.
[0024] In a boost converter, a limitation on the coil / inductor current can be implemented
to prevent inductor saturation, excessive battery currents, or damage in the circuit
itself. When current limitation is active, for example for heavy loads, the target
boost voltage can no longer be reached and the controller is no longer able to control
the boost voltage. There can be a large error voltage on the input of the controller
in this situation. Controllers can have an integrating function for high boost voltage
accuracy, in which case the integrated error voltage will drift away when current
limiting is occurring. When the heavy load is no longer present, the current will
stay at the maximum level until the boost voltage is back at the target level. Then
the recovery starts, the error voltage changes sign and the integrated error voltage
will decrease back to normal operating level. The amount of drift will determine how
much time this recovery takes. This can lead to a much higher boost voltage overshoot
than in the case of a load step under non-limited current levels. The longer the recovery
takes, the higher the overshoot on the boost voltage will be.
[0025] Figure 1 shows an example circuit diagram of a boost converter 100, which provides
an output-voltage
VBST. The boost converter includes an inductor 102 that is connected between a battery
104 and two switches:
Switch 1 106 and
Switch 2 108. The two switches 106, 108 connect the inductor 102 either to ground 110 or to
an output capacitor
CBST 112.
Switch 1 106 is closed with switching frequency
fBST (or period T) and duty cycle
D. Switch 2 108 is closed when
Switch 1 106 is open. The resulting current
IL in the inductor 102 is a triangular shaped current with slopes determined by the
battery voltage
VBAT and the difference between battery and output-voltage
VBAT-
VBST. In the configuration of Figure 1, the ratio between output-voltage and battery voltage
is determined by the duty cycle D used to operate the switches (
D'= 1-D):

[0026] Figure 2 shows a boost converter 200 with peak current mode control. In the current
mode controlled boost converter 200 of Figure 2, the duty cycle control of the switches
206, 208 is performed with peak current mode control. In peak current mode control,
the inductor current
IL 214 is compared to a peak-current-control-signal
IPEAK 218 by a comparator 214.
[0027] At the beginning of each cycle, at the rising edge of
fBST, switch 1 206 is closed and
IL 214 increases with slope
S1 (>0). As soon as
IL 214 is equal to the peak-current-control-signal
IPEAK 218, the comparator 214 resets a latch 216, and
switch 2 208 is closed (and
switch 1 206 is opened). In response, the inductor current
IL 214 decreases with slope
S2 (<0).
[0028] Figure 3 shows a plot of current (on the vertical axis) versus time (on the horizontal
axis) for the inductor current
IL 314 of the circuit shown in Figure 2. As discussed above, Figure 2 shows a boost
converter operating in peak current control. Also shown in Figure 3 is the peak-current-control-signal
IPEAK 318. Furthermore, the switching periods (T) are marked on the horizontal axis, and
a single period is marked up with reference to the duty cycle D that is being used.
[0029] In steady state, the operation shown in Figure 3 defines the relation:

[0030] With this method, the peak value of the triangular inductor current
IL 314 is regulated to the peak-current-control-signal
IPEAK 318. Switch 1 is closed at the rising edge of a clock signal
fBST and opened as soon as the measured current through the coil
IL 314 is equal to the peak-current-control-signal
IPEAK 318. In Figure 3, the average inductor current
IL,AVG 319 is also shown.
[0031] Figure 4 shows the same waveforms as Figure 3, and also an inductor-current-with-error
IL,ERR 420 signal. Figure 4 shows the current waveforms in peak current mode control, including
error propagation but without slope compensation (which will be described below with
reference to Figures 6 to 9).
[0032] IL,ERR 420 is the actual current in the inductor following a change in the load current
(for example at time=0).
IL 414 is the appropriate steady state inductor current for the new load current. As
shown in Figure 4, the system takes a few cycles to adjust to the new load, following
which
IL,ERR 420 is a good match for
IL 414.
IL,AVG 419 represents the average current through the inductor.
[0033] In Figure 4 the propagation of an error
ΔIL,0 422 in the current through the inductor at
t=0 can be seen. The duty cycle
D of the period with
0<
t<
T is altered due to the error in the current
ΔIL,0 422 with:

[0034] The current error
ΔIL,1 at
t=
T will then be equal to:

[0035] The amplification of the current error after one cycle is therefore equal to

[0036] After n cycles, the original error has amplified with a factor
A(n): 
[0037] Since
S2 <0 and
S1 >0, the error has an alternating sign. Note that this system is not stable when |
S2|>|
S1|, so for D >50%. However, in Figure 4, |
S2|<|
S1|, and therefore the magnitude of the error
ΔIL reduces over time.
[0038] Figure 5 shows the same current waveforms in peak current mode control as those of
Figure 4, with corresponding reference numbers in the 500 series, but this time for
error propagation with |S
2|>|S
1|. Figure 5 shows instable behaviour, where the error increases (is amplified) over
time, which leads to undesired sub-harmonic oscillations.
[0039] Figure 6 shows a boost converter 600 with peak current mode control and slope compensation.
[0040] Slope compensation can be used to control the damping and ensure stable behavior
for duty cycles above 50%. With slope compensation, a periodic current
IS 624 with slope
S3 is subtracted from the peak-current-control-signal
IPEAK 618 as shown in Figure 6. The result of the subtraction is (
IPEAK - IS) 625. In this example, linear slope compensation is used; that is,
S3 has a constant value.
[0041] Figure 7 shows the same current waveforms in peak current mode control as those of
Figures 4 and 5, with corresponding reference numbers in the 700 series. Figure 7
also shows (
IPEAK -
IS) 725, which is identified in Figure 6. Figure 7 shows current mode control with slope
compensation and |
S2|>|
S1|, having stable behavior.
[0042] For the circuit of Figure 6 and waveforms of Figure 7, the duty cycle D of the first
period 0<
t<
T will then be altered with

[0043] The current error at
t=
T will now be equal to:

[0044] The amplification of the error is therefore equal to

[0045] After n cycles, the original error
IL,0 has amplified with a factor
A(
n):

[0046] To guarantee stability,
S3 should be chosen in such a way that the magnitude of the term between the brackets
is below 1. Otherwise, the error explodes and sub harmonic oscillations can occur.

[0047] Consider that only
S1 is positive, and
S2 and
S3 are negative. The term between the brackets can be both negative and positive, dependent
on
S3.
[0048] S2 <
S3 < 0 -1 <
A < 0 the
under damped case, the error is alternating in sign.
[0049] S3 < S2 < 0 0 < A < 1 the
over damped case, the error keeps the same sign.
[0050] S3=
S2 < 0
A = 0 the
critically damped case, the error is corrected in a single period.
[0051] The situation with
S3 =
S2, or
A=0, gives the fastest possible transient response. This can be the optimum slope control.
[0052] Figure 8 shows a boost converter 800 with peak current mode control and parabolic
slope compensation. In this example, a periodic current
IS 824 with a parabolic slope is used.
[0053] Figure 9 shows the following current waveforms for the circuit of Figure 8:
- inductor-current IL 914;
- peak-current-control-signal IPEAK 918;
- (IPEAK - IS) 925; and
- average inductor current IL,AVG 919.
[0054] In the boost converter, the current slopes
S1 and
S2 depend on the duty cycle D.

[0055] When a constant value is chosen for
S3 (as in Figures 6 and 7), there is only one operating condition where the damping
is critical (
A=0). Even when |
A|<1 for all operating conditions and stability is guaranteed, this means that the
error correction can take many periods thereby resulting in a slow response. When
optimal slope control with critical damping is required (with
S3 =
S2) for all values of D, then the value of
S3 should also depend on the duty cycle D:

[0056] So, at
t=DT, the slope current derivative should be equal to
S3. Or, in the time domain:

[0057] If the derivative of the slope current
S3 depends linearly on the relative time in the period, the current itself depends on
the square of the relative time, starting at 0 for t/T=0:

[0058] Figure 10 shows a boost converter 1000 with peak current mode control and parabolic
slope compensation, which is similar to the circuit of Figure 8 but with a more detailed
view of the switches and the current measurement. Components that are shown in earlier
figures have been given corresponding reference numbers in the 1000 series, and will
not necessarily be described again here.
[0059] The power switches 1006, 1008 of the boost converter are implemented with power transistors:
MN,POWER 1006, which corresponds to
switch 1 in earlier figures; and
MP,POWER 1008, which corresponds to
switch 2 in earlier figures. These two power switches 1006, 1008 are driven by drivers that
are supplied by voltage sources
VDN 1026 and
VDP 1030 respectively.
[0060] The parabolic slope current
IS 1024 is extracted from the peak-current-control-signal
IPEAK 1018. The resulting current (
IPEAK-
IS) 1025 is forced into a reference transistor M
REF 1030, which is a scaled replica of the power transistor M
N,POWER 1006 with scaling factor
K. At the positive edge of
fBST, M
N,POWER 1006 is switched on by pulling its gate voltage towards
VDN 1026 which is the supply voltage for the driver stage. The gate of the reference
transistor
MREF 1032 is also at
VDN. Sense transistors M
N,SENSE1 1032 and M
N,SENSE2 1034 divide the voltage across the power transistor M
N,POWER 1006 by a factor two when M
N,POWER 1006 is on, and pull the positive input of the comparator 1014 to ground when M
N,POWER 1006 is off.
[0061] The output of the comparator 1014 will now change when:

[0062] When the output of the comparator 1014 changes: the latch 1016 is reset; M
N,POWER 1006 is switched off; and M
P,POWER 1008 is switched on until the next positive edge of
fBST.
[0063] Figure 11 shows a boost converter 1100 with peak current mode control, parabolic
slope compensation, and a P-I controller 1136 for providing an output-voltage control
loop.
[0064] The set point for the output-voltage
VBST in Figure 11 is created by a programmable current source 1137, which provides a current
ISET. The programmable current source 1137 can use
ISET to set the voltage across a sense resistor
RSH 1140. This voltage across
RSH 1140, can be referred to as a sense-voltage
VSH 1138, and is equal to:

[0065] The P-I controller 1136 compares this sense-voltage
VSH 1138 to a reference voltage
VREF 1142, which is provided by a reference-voltage-source in Figure 11. The difference
between
VREF 1142 and
VSH 1138 can be referred to as the error voltage. The P-I controller 1136 then translates
the error voltage to the peak-current-control-signal
IPEAK 1118.
[0066] The P-I controller 1136 adjusts the peak-current-control-signal
IPEAK 1118 (based on the error voltage) to set the output-voltage
VBST of the boost converter. More particularly, the P-I controller 1136 can adjust the
current through the boost converter and bring the sense-voltage
VSH closer to the reference-voltage
VREF. In this example, the peak-current-control-signal
IPEAK 1118 is used to control the duty cycle for the switching cycle of the boost converter,
thereby controlling the output-voltage V
BST.
[0067] Figure 12 shows a boost converter 1200, which is similar to the boost converter of
Figure 11, with peak current mode control and parabolic slope compensation. Figure
12 shows further details of an implementation of the P-I controller 1236, which provides
the output voltage control loop.
[0068] If the boost voltage is on target, then
VREF =
VSH and

[0069] The P-I controller 1236 includes a first transconductance-amplifier-control-block
GM1 1243 and a second transconductance-amplifier-control-block
GM2 1244. The first transconductance-amplifier-control-block
GM1 1243 receives the sense-voltage
VSH 1238 and the reference voltage
VREF 1242, and provides an output signal that has a current that is based on the difference
between
VSH 1238 and
VREF 1242. A control-resistor
R1 1246 and a control-capacitor
C1 are connected in series between the output terminal of the first transconductance-amplifier-control-block
GM1 1243 and ground. In this way, the current that is provided as the output signal of
the first transconductance-amplifier-control-block
GM1 1243 causes a control-voltage
VC 1256 to be dropped across the control-resistor
R1 1246 and the control-capacitor
C1.
[0070] The P-I controller 1236 can generate the control-voltage Vc 1256 by integrating the
difference between: (i) the sense-voltage
VSH 1238; and (ii) the reference-voltage
VREF 1242. In this way, the control-voltage
VC 1256 can be considered as the integrated error voltage in some examples.
[0071] The second transconductance-amplifier-control-block
GM2 1244 converts the control-voltage Vc 1256 to the peak-current-control-signal
IPEAK 1218. An input terminal of the second transconductance-amplifier-control-block
GM2 1244, which receives the control-voltage
VC 1256, can be referred to as a control-voltage-input-terminal.
[0072] Figure 13 shows a boost converter 1300, which is similar to the boost converter of
Figure 12, with peak current mode control, parabolic slope compensation, and an output
voltage control loop. Figure 13 also includes a current-limiter 1350.
[0073] In this example, the output signal from the second transconductance-amplifier-control-block
GM2 1344 is labelled as
ITARGET 1354 and may be referred to as a target-current-control-signal. In the same way as
discussed above, the target-current-control-signal
ITARGET 1354 is configured to adjust the current through the boost converter in order to
bring the sense-voltage V
SH closer to the reference-voltage V
REF; optionally by setting the peak current through the inductor in the boost converter
1300.
[0074] The current-limiter 1350 can limit the peak-current-control-signal
IPEAK 1318. The current-limiter is connected to the output terminal of the second transconductance-amplifier-control-block
GM2 1344, such that it receives the target-current-control-signal
ITARGET 1354. The current-limiter 1350 provides the peak-current-control-signal
IPEAK as the target-current-control-signal
ITARGET 1354 limited to a max-current-control-value
IMAX 1352. That is:
- (i) the peak-current-control-signal IPEAK 1318 equals the target-current-control-signal ITARGET 1354, when the target-current-control-signal ITARGET 1354 is less than or equal to the max-current-control-value IMAX 1352; and
- (ii) the peak-current-control-signal IPEAK 1318 equals the max-current-control-value IMAX 1352, when the target-current-control-signal ITARGET 1354 is greater than the max-current-control-value IMAX 1352.
[0075] The max-current-control-value
IMAX 1352 can be considered as defining a programmable maximum average current level.
The maximum peak current level can depend on the maximum average current
IL,AVG,MAX, battery voltage
VBAT, boost voltage
VBST, inductor value L and switching frequency
f:

[0076] Figure 14 shows an example implementation of a second transconductance-amplifier-control-block
1444, which provides the functionality of both the second transconductance-amplifier-control-block
GM2 and the current-limiter of Figure 13.
[0077] The second transconductance-amplifier-control-block 1444 receives the control-voltage
VC 1456 as an input signal, and provides the peak-current-control-signal
IPEAK 1418 as an output signal. In line with the above discussion, the second transconductance-amplifier-control-block
1444 converts the control-voltage
VC 1456 into a current signal, and then limits / caps that current signal at a maximum
value in order to provide the peak-current-control-signal
IPEAK 1418.
[0078] The second transconductance-amplifier-control-block 1444 includes a max-current-source
1458, which provides a current signal at the max-current-control-value
IMAX 1452. The second transconductance-amplifier-control-block 1444 also includes a first
current mirror 1461, which includes a first-mirror-first-transistor MN3A 1460 and
a first-mirror-second-transistor MN3B 1464. The conduction channel of the first-mirror-first-transistor
MN3A 1460 is connected in series with the max-current-source 1458.
[0079] The second transconductance-amplifier-control-block 1444 also includes an input-transistor
MN1A 1468, which receives the control-voltage
VC 1456 at its control terminal. The conduction channel of the input-transistor MN1A
1468 is connected in series with: (i) a conversion-resistor
RCONV 1466; and (ii) the conduction channel of the first-mirror-second-transistor MN3B
1464. The value of the conversion-resistor
RCONV 1466 defines the value of transconductance of the second transconductance-amplifier-control-block
1444:
GM2 = 1/
RCONV.
[0080] Figure 14 also shows a second current mirror 1462, which includes a second-mirror-first-transistor
MN2A 1470 and a second-mirror-second-transistor MN2B 1472. The conduction channel
of the second-mirror-first-transistor MN2A 1470 is also connected in series with the
conduction channel of the input-transistor MN1A 1468. The conduction channel of the
second-mirror-second-transistor MN2B 1472 provides the peak-current-control-signal
IPEAK 1418.
[0081] (In this example the transistors are FETs, and therefore their control terminals
are gate terminals, and their conduction channels extend between a drain terminal
and a source terminal. In other examples, BJTs can be used instead of FETS.)
[0082] For low levels of the control-voltage
VC 1456, the first-mirror-second-transistor MN3B 1464 is in its linear region, and therefore
the peak-current-control-signal
IPEAK 1418 will be equal to

[0083] That is, changes to the control-voltage Vc 1456 when it has a "low value" (that is,
one that does not saturate the first-mirror-second-transistor MN3B 1464), cause a
change in the current flowing through the second-mirror-first-transistor MN2A 1470,
which in turn causes a change in the
IPEAK current 1418 flowing through the second-mirror-second-transistor MN2B 1472.
[0084] When the control-voltage
VC 1456 is sufficiently increased, the peak-current-control-signal
IPEAK 1418 will saturate to
IMAX 1452 because the first-mirror-second-transistor MN3B 1464 will go into the saturation
region, and limit the current through the input-transistor MN1 1468 to
IMAX 1452. Any further voltage increase of the control-voltage
VC 1456 will be followed by a voltage increase on the drain of the first-mirror-second-transistor
MN3B 1464 without changing the current through the first-mirror-second-transistor
MN3B 1464.
[0085] Therefore, when the load current is increased, a point will be reached where the
current through the coil / inductor of the boost converter will be limited.
[0086] Figure 15 shows a plot of inductor current
IL versus time for a step-up and step-down in the load current, for the circuit of Figure
13. The timing of the load step down is labelled in Figure 15 with reference number
1501. For the plot of Figure 15, the inductor current
IL does not exceed the max-current-control-value
IMAX, and therefore no current limiting occurs.
[0087] Figure 16 shows another plot of inductor current
IL versus time for a step-up and step-down in the load current. Again, the timing of
the load step down is labelled in Figure 16 with reference number 1601. The size of
the load current step for Figure 16 has double the amplitude of the load current step
for Figure 15. For Figure 16, the maximum average inductor current is exceeded and
therefore the inductor current
IL is limited to the max-current-control-value
IMAX.
[0088] Figure 17 shows two plots of output-voltage
VBST versus time:
- a first plot 1703 that corresponds to the load step-up and step-down of Figure 15,
that is a load step that does not result in current limiting; and
- a second plot 1705 that corresponds to the load step-up and step-down of Figure 16,
that is a load step that does result in current limiting.
[0089] The second plot 1705 shows that the output-voltage
VBST can collapse when current limiting is used.
[0090] With reference to the PI-controller 1336 of Figure 13, when current limiting is active,
the error voltage at the input of
GM1 1343 will grow since the control loop is no longer capable of regulating the output-voltage
VBST. The integrating character of the PI-controller 1336 can cause the control-voltage
VC to drift to the supply voltage of
GM1 1343, and away from the normal operating region.
[0091] Figure 18 shows two plots of the control-voltage
VC versus time:
- a first plot 1807 that corresponds to the load step-up and step-down of Figure 15,
that is a load step that does not result in current limiting; and
- a second plot 1809 that corresponds to the load step-up and step-down of Figure 16,
that is a load step that does result in current limiting.
[0092] As can be seen from the second plot 1809 in Figure 18, due to the integrating character
of the PI-controller, the control-voltage
VC drifts away from the normal operating region. When the load current drops to normal
levels again at load step 1801, the current through the coil
IL will stay at the maximum level until the output-voltage
VBST is regulated back to the target level. Then recovery starts, the error voltage (difference
between
VSH and
VREF) changes sign, and the control-voltage Vc will reduce until the normal operating
region is reached again. Because of the drift of
VC during current limiting, this recovery can take a significant amount of time, which
can be too long for some applications. During this recovery time, the coil current
IL can be too high, thereby causing an overshoot on the output-voltage
VBST that is significantly higher than for a load step without current limiting. This
undesired overshoot can be seen in the second plot of Figure 17, after the load step.
[0093] One or more of the following circuits can advantageously prevent or reduce drift
of the control-voltage
VC when current limiting is used. For example, the control-voltage
VC can be clamped to the voltage level at which current limiting starts. In this way,
the control-voltage
VC may not leave the normal operating region. When the load subsequently reduces to
a lower level that does not require current limiting, normal operation can be resumed
more quickly, and in some examples immediately. This can beneficially reduce the overshoot
on the output-voltage to a level that would occur without current limiting.
[0094] Figure 19 shows an example embodiment of a circuit 1900 that includes a switched-mode-power-supply
(SMPS) 1901. The SMPS 1901 may be a boost converter, as discussed above. In other
examples, the SMPS 1901 may be a buck converter, a buck-boost converter or a flyback
converter, as non-limiting examples.
[0095] The SMPS 1901 receives a current-control-signal
ICONTROL 1918, and provides an output-voltage
VOUT based on the current-control-signal
ICONTROL 1918. In some examples the current-control-signal
ICONTROL 1918 can be a peak-current-control-signal
IPEAK, as discussed above.
[0096] The circuit 1900 includes a controller 1936. In some examples, the controller 1936
can operate the SMPS 1901 in a fixed frequency mode of operation. The controller 1936
can generate a control-voltage
VC 1956 based on the difference between: (i) a sense-voltage
VSH 1938; and (ii) a reference-voltage
VREF 1942. The sense-voltage
VSH 1938 is representative of the output-voltage
VOUT of the SMPS 1901, and can be implemented as shown in Figure 11 for example. In this
example, the control-voltage Vc 1956 is a signal that is internal to the controller
1936.
[0097] The controller 1936 can also generate a target-current-control-signal
ITARGET 1954 based on the control-voltage Vc 1956. The target-current-control-signal
ITARGET 1954 is configured to adjust the current through the SMPS 1901 in order to bring
the sense-voltage
VSH 1938 closer to the reference-voltage
VREF 1942. The controller 1936 can perform this functionality in the same way as described
above, for example with reference to Figure 11.
[0098] The circuit 1900 also includes a current-limiter 1950. The current-limiter 1950 can
be implemented as part of the controller 1936 (as will be discussed below with reference
to Figure 20), or as a separate block as shown in Figure 19. The current-limiter 1950
provides the current-control-signal
ICONTROL 1918 as the target-current-control-signal
ITARGET limited to a max-current-control-value
IMAX 1952, in the same way as discussed above.
[0099] The circuit 1900 further includes a clamp-circuit 1974, which is configured to set
the control-voltage
VC 1956 to a clamp-voltage-value when the current-limiter 1950 provides the current-control-signal
ICONTROL having the limited value of the max-current-control-value
IMAX 1952. Optional feedback from the current-limiter 1950 to the clamp-circuit 1974 is
shown schematically as a dashed line in Figure 19. The clamp-circuit 1974 can directly
or indirectly set the control-voltage
VC 1956 to the clamp-voltage-value; for example by affecting / controlling one or more
components in the output voltage control loop, as will be described in more detail
below.
[0100] The clamp-circuit 1974 can advantageously enable the circuit 1900 to quickly return
to a normal operating mode after a heavy load current step in which the current-control-signal
ICONTROL 1918 is limited. This can reduce the drift of the control-voltage
VC 1956 (especially when the controller 1936 includes an integrator). Therefore, the
voltage overshoot in the output-voltage
VOUT of the SMPS 1901 can be reduced. In some examples, the overshoot in the output-voltage
VOUT in the case of a recovery after a current limiting situation is not higher than recovery
in the case without current limiting.
[0101] The clamp-voltage-value may be the value of the control-voltage
VC 1956 when current limiting begins. In this way, if the max-current-control-value
IMAX 1952 is adjustable, then advantageously the associated clamp level can also automatically
vary with this setting. Alternatively, the clamp-voltage-value may be a predetermined
value that enables the controller 1936 to adequately return to a normal operating
mode after current limiting has stopped. If a constant, predetermined value is used,
then in some applications it should be set at a level that corresponds to the level
for the highest possible setting for the max-current-control-value
IMAX 1952. In this way, the clamp-voltage-value should not limit the range for the adjustable
maximum current
IMAX 1952.
[0102] The control-voltage VC 1956 may be:
- (i) based on the difference between (a) the sense-voltage VSH 1938, and (b) the reference-voltage VREF 1942, when the resultant control-voltage VC 1956 would produce a target-current-control-signal ITARGET 1954 that is less than or equal to the max-current-control-value IMAX 1952; and
- (ii) fixed as the clamp-voltage-value when: the difference between (a) the sense-voltage
VSH 1938, and (b) the reference-voltage VREF 1942 would produce a target-current-control-signal ITARGET 1954 that is greater than the max-current-control-value IMAX 1952.
[0103] The example of the controller 1936 shown in Figure 19 includes two sub-processing
blocks 1943, 1944. The first sub-processing block 1943 processes the sense-voltage
VSH 1938 and the reference-voltage
VREF 1942, in order to provide the control-voltage Vc 1956. The second sub-processing
block 1944 processes the control-voltage
VC 1956, in order to generate the target-current-control-signal
ITARGET 1954. One or both of the sub-processing blocks 1943, 1944 may be implemented as transconductance-amplifier-control-blocks.
[0104] Figure 20 shows an example embodiment of a transconductance-amplifier-control-block
2044 that provides the functionality of the second sub-processing block, the current-limiter,
and the clamp-circuit of Figure 19. In this example, the current-control-signal
ICONTROL of Figure 19 is implemented as a peak-current-control-signal
IPEAK 2018
[0105] The transconductance-amplifier-control-block 2044 is configured to receive the control-voltage
VC 2056; prevent the control-voltage
VC 2056 from increasing when the peak-current-control-signal
IPEAK 2018 has the limited value of the max-current-control-value
IMAX 2052; and convert the control-voltage
VC 2056 to the peak-current-control-signal
IPEAK.
[0106] Components of Figure 20 that have already been described with reference to Figure
14 have been given corresponding reference numbers in the 2000 series, and will not
necessarily be described again here.
[0107] The clamp-circuit 2074 includes: a bias-voltage-source 2076; a voltage amplifier,
which in this example is an opamp 2078; and a clamp-transistor MN2 2080. In this example,
the clamp-circuit 2074 also includes an optional clamp-switch MN4 2082 that can be
used to selectively enable or disable the functionality of the clamp-circuit 2074.
[0108] The clamp-circuit 2074 can be considered as a voltage regulator, in that it can regulate
the control-voltage
VC to the clamp-voltage-value, for all values of
VSH and
VREF that would cause
ITARGET to be set to a value that is equal to or greater than the max-current-control-value
IMAX 2052. That is, once the control-voltage
VC (received from a preceding component in the controller) reaches a value where current
limiting starts, any subsequent increases to the control-voltage
VC, that would be expected due to a mismatch between
VSH and
VREF, do not occur because the control-voltage
VC is regulated by the clamp-circuit 2074. In this way, the SMPS is not further controlled
in an attempt to reduce the difference between
VSH and
VREF. This is because the difference between
VSH and
VREF cannot be further reduced due to the current limiting that is performed by the current-limiter.
[0109] The opamp 2078 has a first-opamp-input-terminal and a second-opamp-input-terminal.
The first-opamp-input-terminal receives a bias-voltage-signal
VBIAS from the bias-voltage-source 2076. The bias-voltage-signal
VBIAS may also be referred to as a saturation voltage because it is used to determine whether
or not a transistor has entered a saturation region, as will be discussed below. The
second-opamp-input-terminal receives a current-limited-indicator-signal. The current-limited-indicator-signal
is indicative of whether or not the peak-current-control-signal
IPEAK 2018 is set to the max-current-control-value
IMAX 2052. The opamp has an opamp-output-terminal that is connected to a control terminal
of the clamp-transistor MN2 2080. The conduction channel of the clamp-transistor MN2
2080 is connected in series between the control terminal of the input-transistor MN1
2068 and ground.
[0110] In this example, the second-opamp-input-terminal is connected to the drain of the
first-mirror-second-transistor MN3B 2064. As discussed above with reference to Figure
14, when the peak-current-control-signal
IPEAK 2018 is set to the max-current-control-value
IMAX 2052, the first-mirror-second-transistor MN3B 2064 goes into a saturation region.
When the first-mirror-second-transistor MN3B 2064 is in the saturation region, the
voltage at its drain increases, thereby increasing the voltage at the second-opamp-input-terminal.
The bias-voltage-signal
VBIAS can have a value that corresponds to the drain-source saturation voltage of the first-mirror-second-transistor
MN3B 2064 (
VDS,SAT.MN3B). When the voltage at the second-opamp-input-terminal exceeds the bias-voltage-signal
VBIAS that is received at the first-opamp-input-terminal, the opamp 2078 controls the clamp-transistor
2080 such that the control-voltage Vc at the control terminal of the clamp-transistor
MN2 2080 is clamped at the clamp-voltage-value. This therefore causes the peak-current-control-signal
to have a value of
IMAX.
[0111] In this way, the clamp-circuit 2074 can detect when current limiting is being performed
by comparing the drain voltage of the first-mirror-second-transistor MN3B 2064 with
a bias-voltage that is equal to
VDS,SAT,MN3B. Then, as soon as current limiting occurs, the drain voltage of the first-mirror-second-transistor
MN3B 2064 will exceed
VBIAS, and the opamp 2078 in combination with the clamp-transistor MN2 2080 can regulate
the voltage drop across the first-mirror-second-transistor MN3B 2064 to the saturation-voltage
provided by the bias-voltage-source 2076. That is, the clamp-circuit 2074 can prevent
the control-voltage
VC 2056 from increasing when a transistor (such as the first-mirror-second-transistor
MN3B 2064) is in saturation. In some examples, saturation can be identified by the
voltage drop across the transistor exceeding a saturation-voltage of the transistor.
[0112] In this example, the clamp-circuit 2074 is configured to provide a flag-signal 2086
that is representative of whether or not the control-voltage
VC is set to the clamp-voltage-value (which in this example is the last value of the
control-voltage
VC before current limiting started). To this end, the clamp-circuit 2074 includes a
clamp-comparator 2084 that compares the bias-voltage-signal
VBIAS with the current-limited-indicator-signal, and sets the flag-signal 2086 to a value
that is representative of clamping when the current-limited-indicator-signal is greater
than the bias-voltage-signal
VBIAS.
[0113] In some applications, the clamp-circuit 2074 can pass the flag-signal 2086 to a controllable-block
that is configured to be automatically controlled based on the flag-signal 2086. For
instance, an audio amplifier circuit may include a controllable-block that is configured
to be automatically controlled based on the flag-signal 2086. The controllable-block
may comprise an amplifier, and the amplifier can automatically control an audio level
of an output signal of the audio amplifier circuit based on the flag-signal. The audio
amplifier may be a class-D audio amplifier. For example, the amplifier can automatically
reduce an audio level of the output signal of the audio amplifier circuit when the
flag-signal 2086 indicates that the control-voltage Vc is clamped. This can be on
the basis that voltage clamping occurs when the volume is too high and can be damaging
to the ears of listeners and / or in order to avoid excessive power consumption and
/ or to avoid high distortion of the audio signal when the supply voltage for the
amplifier
VBST is not high enough when clamping occurs.
[0114] In some examples, a cascode (not shown) can be provided in series with the first
current mirror 2061 in order to advantageously increase the accuracy of the first
current mirror 2061.
[0115] Figures 21 to 24 show plots of signals for the circuit of Figure 19, where the SMPS
is implemented as a boost converter in the same way as Figure 13.
[0116] Figure 21 shows three plots of output-voltage
VBST versus time:
- a first plot 2103 that corresponds to a load step-up and step-down that does not result
in current limiting;
- a second plot 2105 that corresponds to a load step-up and step-down for the circuit
of Figure 13, that is a load step that does result in current limiting but is for
a circuit that does not have voltage clamping; and
- a third plot 2106 that corresponds to a load step-up and step-down for the circuit
of Figure 19, that is a load step that does result in current limiting and is for
a circuit that does have voltage clamping.
[0117] A comparison of the second and third plots 2105, 2106 shows that use of the control-voltage
clamping advantageously reduces the overshoot in the output-voltage
VBST.
[0118] Figure 22 shows the inductor current
IL during a load step-up and step-down, where both current limiting and voltage clamping
are used.
[0119] Figure 23 shows three plots of the control-voltage
VC versus time:
- a first plot 2307 that correspond to a load step-up and step-down that does not result
in current limiting; and
- a second plot 2309 that corresponds to a load step-up and step-down for the circuit
of Figure 13, that is a load step that does result in current limiting but is for
a circuit that does not have voltage clamping; and
- a third plot 2310 that corresponds to a load step-up and step-down for the circuit
of Figure 19, that is a load step that does result in current limiting and is for
a circuit that does have voltage clamping.
[0120] Figure 23 shows that the control-voltage
VC does not significantly drift away from its normal operation value while the current
is being limited.
[0121] Figure 24 shows a plot of the flag-signal that is representative of whether or not
the control-voltage Vc is set to the clamp-voltage-value.
[0122] Figure 25 shows an example embodiment of the transconductance-amplifier-control-block
of Figure 20, which provides the functionality of the second sub-processing block,
the current-limiter, and the clamp-circuit of Figure 19. In particular, further details
of the clamp-circuit 2574 are included in Figure 25. The opamp of Figure 20 is constructed
with a differential matched pair (MP3A and MP3B) and two bias current sources. The
comparator in Figure 20, which generates the flag-signal 2586, is built with: a transistor
MN5, a bias current source and an inverter.
[0123] Figure 25 also shows the bias-voltage-source 2576 of Figure 20.
[0124] Figure 26 shows an example embodiment of a transconductance-amplifier-control-block,
which is similar to that of Figure 25. In Figure 26, the functionality of the voltage
source
VBIAS of Figure 25 has been replaced by a systematic offset on the input of the opamp by
choosing different sizes of the input-transistors MP3A 2692 and MP3B 2690.
[0125] Figure 27 shows another example embodiment of a circuit 2700 that includes a SMPS
2701 and a controller 2736. In this example, the SMPS 2701 is a boost converter. Features
of Figure 27 that are shown in either of Figures 13 or 19 have been given corresponding
reference numbers in the 2700 series, and will not necessarily be described again
here.
[0126] The controller 2736 incorporates the functionality of a current-limiter 2750 in the
same way as discussed above. The controller 2736 also incorporates the functionality
of a clamp-circuit, although in this example the clamp-circuit indirectly sets the
control-voltage
VC 2756 to a clamp-voltage-value. As will be discussed in more detail with reference
to Figure 28, the controller 2736 can provide a non-zero over-current-signal
IOC 2794 when the current-control-signal
IPEAK 2718 has the limited value of
IMAX 2752. The over-current-signal
IOC 2794 is configured to increase
VSH independently of the current through the boost converter 2701, thereby effectively
lowering the set-point for the boost voltage to a different level that is low enough
to maintain the load current with
IPEAK=
IMAX.
[0127] The boost converter 2701 includes: a sense-resistor
RSH 2740, which is configured to provide the sense-voltage
VSH 2738 to the controller 2736; and a programmable current source 2737 which provides
a current
ISET through the sense-resistor
RSH 2740. In this example, because the over-current-signal
IOC 2794 is also provided to the sense-resistor
RSH 2740, the sense-voltage
VSH 2738 is equal to:

where

[0128] Therefore, a non-zero, positive, value for the over-current-signal
IOC 2794 effectively decreases
VBST,TARGET with
RSH*
IOCP which causes
VBST 2738 to decrease while keeping
VSH constant and also keeping the error voltage
VREF-
VSH constant.
VBST,TARGET can be referred to as a target-output-voltage.
[0129] In Figure 27, the voltage across the sense resistor
RSH 2740 is clamped using the over-current-signal
IOC 2794. This has the knock-on effect of (indirectly) clamping the control-voltage
VC 2756. Under normal operating conditions, when the target-current-control-signal
ITARGET 1954 is below the maximum current level and therefore is not being limited, the over-current-signal
IOC 2794 is zero. However, when current limiting occurs, the over-current-signal
IOC 2794 will increase to limit the voltage across the sense resistor
RSH 2740. The net effect is that the target boost voltage
VBST,TARGET is reduced with
RSH*
IOCP. Instead of regulating the boost voltage using the inductor current
IL 2714 as is done under normal operating conditions, now the inductor current
IL 2714 is regulated to the maximum level using the boost voltage. Since the loop provided
by the controller 2736 is still in control, the error voltage will be small and the
control-voltage (integrated error voltage)
VC 2756 will not drift away.
[0130] In the same way as Figure 19, the controller 2736 of Figure 27 includes a second
sub-processing block 2744 that processes the control-voltage
VC 2756 in order to generate the target-current-control-signal
ITARGET 2754.
[0131] Figure 28 shows an example implementation of the second sub-processing block of Figure
27, which in this example is a transconductance-amplifier-control-block 2844. Features
of the transconductance-amplifier-control-block 2844 that have already been described
with reference to Figure 14, 20, 25 or 26 will not necessarily be described again
here.
[0132] The transconductance-amplifier-control-block 2844 incorporates the functionality
of the current-limiter and the clamp-circuit such that the transconductance-amplifier-control-block
2844 is configured to: receive the control-voltage
VC 2856, and provide a non-zero over-current-signal
IOC 2894 when the current-control-signal
IPEAK 2518 has the limited value of I
MAX 2852. As discussed above, the over-current-signal
IOC 2894 is configured to change the boost voltage
VBST,TARGET independently of the current through the switched-mode-power-supply. The transconductance-amplifier-control-block
2844 also converts the control-voltage
VC 2856 to the current-control-signal
IPEAK 2818.
[0133] One or more of the examples disclosed herein can be used with any current mode controlled
DC-DC converter, and especially those with an integrating voltage control loop.
[0134] The instructions and/or flowchart steps in the above figures can be executed in any
order, unless a specific order is explicitly stated. Also, those skilled in the art
will recognize that while one example set of instructions/method has been discussed,
the material in this specification can be combined in a variety of ways to yield other
examples as well, and are to be understood within a context provided by this detailed
description.
[0135] In some example embodiments the set of instructions/method steps described above
are implemented as functional and software instructions embodied as a set of executable
instructions which are effected on a computer or machine which is programmed with
and controlled by said executable instructions. Such instructions are loaded for execution
on a processor (such as one or more CPUs). The term processor includes microprocessors,
microcontrollers, processor modules or subsystems (including one or more microprocessors
or microcontrollers), or other control or computing devices. A processor can refer
to a single component or to plural components.
[0136] In other examples, the set of instructions/methods illustrated herein and data and
instructions associated therewith are stored in respective storage devices, which
are implemented as one or more non-transient machine or computer-readable or computer-usable
storage media or mediums. Such computer-readable or computer usable storage medium
or media is (are) considered to be part of an article (or article of manufacture).
An article or article of manufacture can refer to any manufactured single component
or multiple components. The non-transient machine or computer usable media or mediums
as defined herein excludes signals, but such media or mediums may be capable of receiving
and processing information from signals and/or other transient mediums.
[0137] Example embodiments of the material discussed in this specification can be implemented
in whole or in part through network, computer, or data based devices and/or services.
These may include cloud, internet, intranet, mobile, desktop, processor, look-up table,
microcontroller, consumer equipment, infrastructure, or other enabling devices and
services. As may be used herein and in the claims, the following non-exclusive definitions
are provided.
[0138] In one example, one or more instructions or steps discussed herein are automated.
The terms automated or automatically (and like variations thereof) mean controlled
operation of an apparatus, system, and/or process using computers and/or mechanical/electrical
devices without the necessity of human intervention, observation, effort and/or decision.
[0139] It will be appreciated that any components said to be coupled may be coupled or connected
either directly or indirectly. In the case of indirect coupling, additional components
may be located between the two components that are said to be coupled.
[0140] In this specification, example embodiments have been presented in terms of a selected
set of details. However, a person of ordinary skill in the art would understand that
many other example embodiments may be practiced which include a different selected
set of these details. It is intended that the following claims cover all possible
example embodiments.